This drug has already failed to be effective, it's extremely expensive, and it has bad side effects that require periodic MRIs to check for brain inflammation.

This drug has already failed to be effective, it's extremely expensive, and it has bad side effects that require periodic MRIs to check for brain inflammation. If the amyloid plaque hypothesis is wrong (which it almost certainly is based on the failure of other drugs) the destruction if amyloid plaques is likely to be harmful to the brain with no benefit. Looking this as a 1% chance of a cure is not really accurate because if this drug was really capable of completely curing Alzheimer's there is no way the existing trials wouldn't have found that. The article also describes additional reasons why it is generally bad to approve drugs that most likely don't actually work.